3084results about "Web data querying" patented technology

A system is disclosed that uses expert systems to monitor and evaluate quality in a large language model. The system can include a prompt library that associates prompts with areas of expertise. The system selects prompts from the library and evaluates first responses generated by a large language model for the set of prompts against second responses generated by a modified version of the large language model to the set of prompts. The evaluation uses expert systems associated with the areas of expertise for the set of prompts. If the system determines that the evaluation indicates a degradation criterion is met, the system may take remedial action. The system provides an effective way to evaluate and prevent the use of modified language models that do not meet the required standards.

A system and method for extracting data from large language model (LLM) outputs, including: training a model using labeled data items to assign rankings to LLMs; selecting, by the trained model, one or more of the LLMs based on the rankings; sending an LLM prompt to selected models; and outputting, by the model, a refined response to the prompt based on responses to the prompt by the LLMs. Some LLM prompts according to some embodiments may include different sets input parameters of different types—such as, e.g., a set of block parameters and a set of editorial parameters. In some embodiments, a model or LLM may be updated or retrained using a reinforcement learning approach and based output items or refined responses generated by that model or LLM—which may for example be scored or ranked and used in combination with reward or cost functions to update model parameters.

The present disclosure relates to a system and method for generating an outcome in response to a user query using an AI agent controller equipped with a large language model. The AI agent controller receives the user query, determines a sequence of tasks required to achieve the outcome, and communicates the sequence to a Central Agent Name and Attribute Registry (CANAR) containing a repository of registered AI agents. The CANAR identifies and obtains responses from AI agents capable of executing the tasks. The AI agent controller then coordinates the execution of these tasks by the selected AI agents to generate the specified outcome. The system further includes functionalities for validating user queries, cataloging AI agents, and dynamically prioritizing tasks based on user preferences or urgency. Additionally, the system supports real-time updates of the AI agent repository and task distribution based on geographical proximity, ensuring efficient and accurate task execution.

A method includes generating a new user query embedding for a new user query received from a user, obtaining an indexed user query matching the new user query from a search engine index, a vector index corresponding to the indexed user query, and a relevancy score corresponding to the indexed user query. The method further includes selecting a vector structure corresponding to the vector index from a plurality of vector structures in a vector store, obtaining, from the vector structure, a result embedding matching the new user query embedding, transmitting, by a user query answer service to an answer generation model, the result embedding and receiving, by the user query answer service, an answer to the new user query from the answer generation model.

The invention provides an implementation method, system and equipment of a commercial intelligent system based on a large model ReAct, and belongs to the technical field of artificial intelligence. The method is applied to a ChatBI platform. Comprising the steps that a main agent receives query request information from a user terminal and guides an LLM to conduct cyclic reasoning on the query request information according to a preset ReAct normal form through a preset Prompt project, and the cyclic reasoning process comprises the steps that 1, input information from an upstream task is analyzed through the LLM, and a to-be-executed action corresponding to a downstream task is determined; step 2, according to each action task corresponding to the action to be executed and the action task type thereof, executing a reasoning action; 3, receiving an inference action feedback result, updating a loop inference context, and determining whether to trigger a next round of loop inference or not until a preset loop termination condition is met; and based on each decision analysis result obtained by loop reasoning, generating a query result and sending the query result to the user terminal.

An application server or other processing entity may receive, via a cloud-based platform, user input that may include at least one request for data. The application server may classify the user input into a first of a plurality of deterministic-stochastic spectrum classifications based on the user input and a probability of mapping the at least one request for data to at least one data location. The application server may retrieve the data from the at least one data location and based on the first deterministic-stochastic spectrum classification. The application server may transmit, based on the first deterministic-stochastic spectrum classification and the user input, an input to a large language model. The application server may present a response to the user input, where the response is based on a combination of an output of the large language model and the data retrieved from the at least one data location.

The invention relates to an industrial equipment maintenance intelligent question-answering system based on multi-agent collaboration. Wherein the input unit is used for receiving text, voice, image or equipment scanning and other multi-mode user input information and analyzing the information into structured problem information; the scheduling unit performs semantic understanding and problem classification on the structured problem information based on the fine-tuned cross-language pre-training language model and an incremental training mechanism; the processing unit calls a corresponding domain agent according to the classification result, and generates an intelligent question and answer processing result including predictive maintenance suggestions, structured reply content and semantic annotation information; and the fusion unit fuses the local knowledge base, the graph database and the networking retrieval information, performs multi-hop semantic reasoning on the intelligent question and answer processing result, and generates multi-modal reply information including text description, image screenshots, prediction curves and recommendation links. The system can support multi-language and multi-mode intelligent question answering and predictive maintenance in a complex industrial maintenance scene.

A large language model (LLM) is used to broaden a search for supporting material for a media project. The LLM is provided with contextual material from the media project and optional grounding material and generates a list of types of supporting material. The list is provided to a machine-learning-based encoder, which encodes the list items into embedding space vectors. A body of source material, which may multimodal, is also encoded into the embedding space. A search engine is used to locate items of source material having embedded space vectors closest to vectors corresponding to the list of types of supporting material. Located items are provided to a media editing application as suggested supporting material. The body of source material may include live external sources. A small language model fine-tuned with a training data set generated by a LLM may be used instead of the LLM.

A method and an electronic device for generating article content are disclosed. The method includes: constructing a subject keyword library by collecting industry trend data, and collecting content materials for article generation; in response to a request for generating an industry information article, determining a target subject keyword and a prompt word text for dialogue with an artificial intelligence AI large language model, to facilitate a generation of an article subject; screening multiple segments of target content materials from the content materials; constructing a prompt word text for dialogue with the AI large language model according to the article subject and the target content materials to generate an article outline, the article outline including multiple sub-subjects; and calling the AI large language model multiple times to generate corresponding text contents for the multiple sub-subjects respectively.

In some embodiments, a method receives query content that includes first visual data and select reference content that includes second visual data. A first visual feature mapping for the query content is generated using a model that is trained to generate visual feature mappings using content pairs where alterations in one instance of the content pair are used in the training. A second visual feature mapping is determined for the reference content. The second visual feature mapping is determined using the model. The method compares the first visual feature mapping and the second visual feature mapping to generate a score that rates a similarity between the query content and the reference content. The score is based on recognizing an alteration in one instance of the query content and the reference content. The score is output.

The invention discloses an intelligent question and answer method and system, electronic equipment and an intelligent question and answer large model, and relates to the technical field of intelligent question and answer. The method comprises the steps of receiving query information reported by a user terminal, and obtaining a user intention according to the query information; if the user intention is the preset high-frequency question and answer intention, obtaining a pre-generated high-frequency full-text question answer corresponding to the user intention, and sending the high-frequency full-text question answer to the user terminal; if the user intention is not the preset high-frequency question and answer intention, performing multi-source heterogeneous knowledge retrieval, generating answer information according to a retrieval result, and sending the answer information to the user terminal, comprising at least one of retrieval in a pre-generated knowledge base according to the query information and network retrieval according to the query information.

The invention discloses an intelligent news broadcasting method and system based on artificial intelligence, and the method comprises the following steps: S1, obtaining data through a news crawling tool, and generating structured event data; s2, generating an event knowledge graph by using an improved BERT model, and performing clustering analysis by using an improved K-Medoids clustering algorithm; s3, generating a dynamic broadcast script through the multi-modal generative adversarial network; s4, analyzing the emotional state of the user, and generating personalized voice broadcast content; s5, generating an enhanced news abstract in combination with an event-driven mechanism and a time sensitivity optimization rule; s6, distributing dynamic weights by using a hierarchical attention mechanism, and dynamically adjusting a news broadcasting sequence in combination with a time sensitivity optimization rule; s7, performing logic consistency detection on the news broadcast content; and S8, adjusting the broadcast content and generating updated news broadcast content in real time. According to the method, personalized news broadcasting is realized by combining an improved algorithm, sentiment analysis, a dynamic broadcasting technology and the like.

The invention discloses a science and technology fast message intelligence perception system based on a large language model, and belongs to the field of intelligence perception, the system comprises an information collection module, a data standardization module and a database, the information collection module obtains multi-mode science and technology fast message data; the data standardization module is used for receiving the multi-modal data output by the information collection module and executing data cleaning, structured processing and semantic standardization operation; after standardization processing, storing the data into a database; the system further comprises an intelligence value screening module, an evaluation system module and a visualization module. And the intelligence value screening module is combined with the evaluation system to identify the intelligence value, and multi-dimensional visual presentation is realized through the visualization module. The method improves the efficiency and accuracy of information recognition and perception, and has significant advantages in the aspects of processing mass information and enhancing the information perception ability.

The invention relates to an internet big data extraction method and device, equipment and a storage medium, and the method comprises the following steps: carrying out distributed crawler collection on an internet data source, obtaining original network data, and converting the original network data into a structured data matrix; and performing multi-level semantic analysis on the matrix, constructing a semantic feature map, and performing topic segmentation and classification to form a topic domain knowledge tree. Association rules in the knowledge tree are further mined, and an implicit knowledge network is constructed. Performing semantic decomposition and expansion on query conditions based on the network to generate expanded query data, and performing similarity matching with the knowledge network to obtain a candidate data set; and finally, performing multi-factor sorting and extraction on the candidate data, and outputting target data, thereby solving the technical problem that in the second-hand car market, due to wide data sources and fuzzy semantics, an existing system has relatively large deviation when performing price prediction and maintenance cost analysis.

A system and method for chatbot and search engine integration comprising traversing a conversational flow of a chatbot, generating a data structure based on the conversational flow, embedding the data structure in a webpage, and directing a search engine to the webpage for the search engine to assign a webpage rank to the data structure. In some embodiments, this integration may allow search engine users to be redirected to a non-public website hosting the chatbot when that chatbot would not otherwise be identified by the search engine.

A search system may receive a search query for documents stored in a database associated with an organization. A search system may retrieve, from the database, search results (including responsive documents) that are responsive to the search query. A search system may initiate display of the search results on the user interface. A search system may transmit a prompt to a large language model, where the prompt includes the search query and information from at least a portion of the responsive documents. A search system may receive a model response with textual data that responds to the search query, where the textual data is generated by the large language model using at least the portion of the responsive documents. A search system may initiate display of the model response in a user interface.

The invention provides a text search method and system based on vector retrieval and large model optimization, and relates to the technical field of text search, and the method comprises the following steps: obtaining a to-be-retrieved text and generating a semantic vector; retrieving similar texts in a vector database based on the semantic vectors; calculating the correlation of the text pairs by using a cross encoder model and sorting the text pairs; and extracting key entities by adopting a knowledge graph optimization model, analyzing an entity relationship, and reordering and filtering search results. According to the method, through semantic vector retrieval, cross coding calculation of correlation and knowledge graph optimization, the accuracy and correlation of text search are improved, user intentions can be better understood, and more accurate search results can be returned.

The invention discloses a public opinion risk dynamic early-warning method and system based on a large language model, and the method and system carry out the semantic recognition, event extraction and situational reasoning of unstructured public opinion data through introducing an advanced large language model, and carry out the dynamic early-warning of the public opinion risk in combination with structured emotion indexes and market dynamics. Potential risk factors in social public opinions can be automatically identified, and risk quantitative evaluation is realized in combination with semantic tags, market behavior data and emotion intensity. The system has strong prospective perception ability and event evolution reasoning ability, can provide rapid and accurate early warning information, and further can generate personalized suggestions and decisions in combination with user risk preferences. The method can be widely applied to a plurality of public opinion sensitive fields such as enterprise brand management, government supervision, financial institutions and media transmission control, and the automation and precision capability of dealing with sudden public opinion risks is effectively improved.

The present application relates to the technical field of electronics, and provides a method for generating a commodity label on the basis of a large language model, and an electronic device. The method specifically comprises: acquiring commodity description information related to a target commodity; performing structured processing on the commodity description information to obtain an objective property feature of the target commodity; and using a fine-tuned large language model to process the objective property feature on the basis of a language prompt lexicon to obtain one or more commodity labels of the target commodity. According to the method, a commodity selling point label can be generated by means of the commodity description information acquired from multiple sources and multiple paths, the source of the commodity description information is expanded, commodity information of different sources can be identified and acquired by means of image and text details, etc., which can assist the model in better understanding the commodity, and generating a more reliable selling point label having a higher confidence level and greater traceability.

The invention belongs to the field of combination of a large model and a knowledge graph, and particularly relates to a large model and knowledge graph combined reasoning method for false news detection, which breaks through the limitation of a traditional detection method in the aspects of knowledge instantaneity and reasoning controllability through heterogeneous architecture design. Based on multi-dimensional features such as timeliness, propagation mode and information density of news propagation, a double-engine-driven intelligent reasoning system is constructed, on one hand, important entity extraction and semantic understanding are carried out through a large language model, and on the other hand, structured fact verification is provided through a dynamic knowledge graph. And the inherent factual illusion problem of a large language model is effectively solved. Compared with a traditional knowledge iteration mode depending on single model parameter updating, the method supports an incremental updating strategy based on knowledge graph nodes.

The invention discloses a method and device for achieving intelligent AI webpage content obtaining based on LLM, a storage medium and a program product, and relates to the technical field of data processing. According to the method, the search instruction input by the user is received, the search instruction is input into the trained semantic analysis model, after the semantic analysis model converts the search instruction into the query parameter, the query parameter is sent to the search engine, then the webpage obtained by the search engine according to the query parameter is received, and the search result is obtained. The method comprises the following steps: acquiring a webpage, determining a target webpage from the webpage according to a preset screening rule, crawling an HTML (Hypertext Markup Language) source code of the target webpage based on a pre-acquired authorization result, inputting the HTML source code into a trained LLM model, and determining a target code label in the LLM model according to the HTML source code to obtain text content in the target code label. Wherein the LLM model has a strong language understanding capability and can dynamically adapt to the change of a webpage structure, so that the time cost of crawling webpage contents is reduced.

Aspects of the disclosed technology include computer-implemented systems and methods for moderating generative content produced by machine-learned generative models and providing outputs indicative of query and response validity for a target domain. The system can generate a query association based on a user query and domain artifacts of a target domain and generate a response association based on generative content and the domain artifacts. The system can embed the domain artifacts in a vector embedding space and / or train a machine-learned model using the domain artifacts. The system can determine the validity of queries and responses for the target domain using the vector embeddings and the machine-learned model. The system can moderate content, such as by filtering queries and / or generative content that does not satisfy association criteria for a target domain. The system can generate outputs such as visualizations of query and response distance relative to a target domain.

The invention discloses a high-value information mining method based on an expert thinking chain large model agent, which comprises the following steps of: inputting a name and a summary of a task, extracting entity and event information keywords in the task through a large language model, and performing text vectorization processing on the keywords and an expert thinking chain template; the cosine distance similarity between the keyword and the expert thinking chain template vector is calculated, and the expert thinking chain template with the highest similarity is obtained through matching; calling a search engine to search entity and event information keywords extracted from the task name and the summary, associating the keywords and searching related contents by utilizing a large model and the search engine aiming at each information keyword, and obtaining and recording a title, a publishing mechanism, publishing time, contents and a website of the searched webpage; extracting recorded titles, publishing structures and contents to obtain an entity, relationship and event triple, and organizing into a public security domain knowledge system for storage; performing intelligent analysis on the constructed public security domain knowledge system according to the expert thinking chain template to obtain a visual statistical chart and a structured analysis conclusion; and according to the visual statistical chart and the structured analysis conclusion, an analysis report is written and generated, and a final information mining analysis report is obtained. According to the method, a large number of complex data sets can be quickly identified, classified and analyzed, so that the process of information extraction and knowledge discovery is accelerated.

The invention discloses a data acquisition method and device, electronic equipment and a storage medium, and relates to the technical field of data processing. The data acquisition method is applied to the data acquisition system, and comprises the following steps: acquiring a user query instruction through terminal equipment, and sending the user query instruction to a vehicle based on a cloud server; the method comprises the following steps: acquiring charging information, riding related information and vehicle position information through a vehicle based on a user query instruction; target display information is determined through a vehicle according to charging information and riding related information, and the target display information and vehicle position information are fed back to a cloud server; and sending the target display information and the vehicle position information to the terminal equipment through the cloud server, so that the terminal equipment displays the target display information and the vehicle position information. According to the embodiment of the invention, remote monitoring of the terminal equipment on the vehicle is realized, the target display information and the vehicle position information are acquired in time, possible faults of the vehicle can be found in time, and the use experience of a user is improved.

In some implementations, the techniques described herein relate to a method including: loading, by a processor, a predictive model, the predictive model including a first set of hidden layers; loading, by the processor, a semantic model, the semantic model including a second set of hidden layers; generating, by the processor, a tenant model using the first set of hidden layers and a third set of hidden layers, the third set of hidden layers receiving, as input, an output of the first set of hidden layers; loading, by the processor, a tenanted training data set; training, by the processor, the tenant model by biasing the first set of hidden layers with the second set of hidden layers and training weights of at least the third set of hidden layers using the tenanted training data set; and building, by the processor, an embedding index using the tenant model.

The invention relates to the technical field of auditing, and provides a multi-modal auditing method based on a large model, which ensures the quality and consistency of data through comprehensive data acquisition and preprocessing, performs feature extraction on image, text and numerical data by using CNN, NLP and other technologies, realizes deep fusion and analysis of multi-modal data, and improves the auditing efficiency. Through doubtful point mining, knowledge graph correlation analysis and sentiment analysis, abnormal modes and potential relationships in data and application of a causal traceability model are deeply revealed, causal relationships and influence factors in a specific scene are further clarified, and finally, the generated audit report is detailed in content, clear in structure and high in practicability. According to the technical scheme, the persuasion of the report is enhanced through data visualization, the auditing accuracy and efficiency are improved through the method, powerful support is provided for auditing decision making, and the problems of insufficient accuracy and low efficiency in the prior art are solved through the technical scheme.

Techniques are generally described for modeling seasonal relevance for online search. A first query or item comprising text is received. A language model may be used to generate a plurality of token embeddings representing the text. In some cases, the first neural network may predict a seasonal relevance vector for the first query or item based on the plurality of token embeddings, the seasonal relevance vector being predicted based at least in part on a similarity between the text of the first query or item and at least one other query / item for which monthly concentration data is available. The seasonal relevance vector, including a respective seasonal relevance score for each month of a year, may be output.

An offline collection system comprises a pipeline for storing metadata of user interface (UI) elements at web pages of a web application. The pipeline comprises crawling uniform resource locators (URLs) of web pages of the web application for content and rendering screenshots of the web pages. The pipeline then prompts a multimodal large language model (LLM) to generate database entries for the web pages comprising UI element metadata derived from the crawled content and rendered screenshots. A response system receives user queries to navigate the web application and augments prompts to an LLM to respond to the user queries with metadata of UI elements relevant to the user queries stored by the offline collection system.

The present disclosure is related to framework for automatically and efficiently finding machine learning (ML) architectures that are optimized to one or more specified performance metrics and / or hardware platforms. This framework provides ML architectures that are applicable to specified ML domains and are optimized for specified hardware platforms in significantly less time than could be done manually and in less time than existing ML model searching techniques. Furthermore, a user interface is provided that allows a user to search for different ML architectures based on modified search parameters, such as different hardware platform aspects and / or performance metrics. Other embodiments may be described and / or claimed.